CN107329831A - A kind of artificial resource dispatching method based on improved adaptive GA-IAGA - Google Patents
A kind of artificial resource dispatching method based on improved adaptive GA-IAGA Download PDFInfo
- Publication number
- CN107329831A CN107329831A CN201710512814.7A CN201710512814A CN107329831A CN 107329831 A CN107329831 A CN 107329831A CN 201710512814 A CN201710512814 A CN 201710512814A CN 107329831 A CN107329831 A CN 107329831A
- Authority
- CN
- China
- Prior art keywords
- individual
- population
- father
- adaptive value
- adaptive
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/12—Digital output to print unit, e.g. line printer, chain printer
- G06F3/1201—Dedicated interfaces to print systems
- G06F3/1223—Dedicated interfaces to print systems specifically adapted to use a particular technique
- G06F3/1237—Print job management
- G06F3/126—Job scheduling, e.g. queuing, determine appropriate device
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/5038—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration
Abstract
The invention discloses a kind of artificial resource dispatching method based on improved adaptive GA-IAGA, methods described includes:S1:The population of initialization generation simulation model and artificial resource;S2:The individual fitness of each individual in population is calculated based on fitness function;S3:Crossover operation is carried out to population based on single-point cross method;S4:Mutation operation is carried out to population based on adaptive mutation probability;S5:Selection operation is carried out to population using based on accumulated probability improved selection opertor;S6:S1 S5 are repeated, reaches and terminates after predetermined step number, the present invention to artificial resource scheduling to based on genetic algorithm, optimizing, improving high flux artificial resource dispatching efficiency, improve the execution efficiency of artificial tasks and the throughput of analogue system.
Description
Technical field
The present invention relates to artificial resource scheduling field.More particularly, to a kind of emulation money based on improved adaptive GA-IAGA
Source dispatching method.
Background technology
The characteristics of there is multi-user's high concurrent due to high-performance high flux Simulation Application, when multiple artificial tasks compete limited
System resource when, efficient task and resource regulating method can greatly improve the throughput of system.Deposit at present
In many task scheduling system and algorithm, but it is not the professional scheduling system for high-performance high flux design of Simulation,
The characteristics of versatility of system can not be directed entirely to Simulation Application carries out special optimization, it is therefore desirable to which research is high in high-performance
The method for scheduling task of Simulation-Oriented application in flux computing system, the performance and entirety of maximized raising Simulation Application are gulped down
Tell rate.
Genetic algorithm is the general framework for solving complex combination optimization problem, and the specific field with problem is unrelated, searches for
Journey without the concern for problem inwardness, only consider fitness function, have been widely used for function optimization, production scheduling,
The fields such as image procossing, machine learning.But for high flux emulation in the model running time it is unknown in the case of, operation heredity calculate
Method solves both to have considered the calculating consumption of model during the model resource mapping with dependence the communication consumption it is further contemplated that between model
When, it is necessary to propose the improvement to genetic algorithm, allow genetic algorithm to be preferably adapted to Mission Scheduling and improve genetic algorithm
Superiority.
Accordingly, it is desirable to provide a kind of artificial resource dispatching method based on improved adaptive GA-IAGA, improves high flux emulation money
Source dispatching efficiency, improves the throughput that task completes efficiency and analogue system.
The content of the invention
It is an object of the present invention to provide a kind of artificial resource dispatching method based on improved adaptive GA-IAGA, to improve
High flux artificial resource dispatching efficiency, improves the throughput that task completes efficiency and analogue system.
To reach above-mentioned purpose, the present invention uses following technical proposals:
The invention discloses a kind of artificial resource dispatching method based on improved adaptive GA-IAGA, methods described includes:
S1:The population of initialization generation simulation model and artificial resource;
S2:The individual fitness of each individual in population is calculated based on fitness function;
S3:Crossover operation is carried out to population based on single-point cross method;
S4:Mutation operation is carried out to population based on adaptive mutation probability;
S5:Selection operation is carried out to population using based on accumulated probability improved selection opertor;
S6:S1-S5 is repeated, reaches and terminates after predetermined step number.
Preferably, the S1 initializes the population with reference to principle and optimization distribution principle is randomly assigned;
The optimization distribution principle will be distributed to different artificial resources with mutually level simulation model, and will be imitative
The height of true mode and the computing capability of artificial resource match.
Preferably, the fitness function is represented with laod unbalance amount.
Preferably, the fitness function is the inverse of laod unbalance amount.
Preferably, the single-point cross method is
The high individual of fitness, equiprobable selection simulation model and person's artificial resource carry out crossover operation;
The low individual of fitness, simulation model and person's artificial resource all carry out crossover operation;
The simulation model and person's artificial resource of coding use different cross methods.
Preferably, the S3 includes:
S31:Two individuals are randomly choosed from population as father's individual;
S32:The adaptive value of two father's individuals is calculated, wherein larger father's individual is compared with planting group mean adaptive value;
S33:If the adaptive value of larger father's individual is less than kind of a group mean adaptive value, enter according to the probability of single-point cross method
The crossover operation of row simulation model and artificial resource, or father's individual is copied directly in population of new generation;
S34:If the adaptive value of larger father's individual is more than kind of a group mean adaptive value, enter according to the probability of single-point cross method
The crossover operation of the equiprobable selection simulation model of row crossover operation, wherein crossover operation and artificial resource, or the father is individual
Body is copied directly in population of new generation;
S35:Described two father's individuals are deleted, S31-S34 is repeated, when the individual number in previous generation populations is 0, intersected
Operation terminates.
Preferably, the adaptive mutation probability is
When population each fitness reaches unanimity or tends to local optimum, mutation probability becomes big;
When colony's fitness is more dispersed, mutation probability diminishes;
Obtain individual higher than the average adaptive value of colony for fitness value, take less mutation probability;
Obtain individual less than the average adaptive value of colony for fitness value, take larger mutation probability.
Preferably, the S4 includes:
S41:An individual is randomly choosed from population as father's individual;
S42:The adaptive value of father's individual is calculated, and is compared with kind of a group mean adaptive value;
S43:If the adaptive value of father's individual is less than kind of a group mean adaptive value, emulation mould is carried out according to adaptive mutation probability
The mutation operation of type and artificial resource, or father's individual is copied directly in population of new generation;
S44:If the adaptive value of father's individual is more than kind of a group mean adaptive value, row variation behaviour is entered according to adaptive mutation probability
Make, wherein the equiprobable selection simulation model of mutation operation and artificial resource carry out mutation operation, or by father individual directly
Copy in population of new generation;
S45:Father's individual is deleted, S41-S44 is repeated, when the individual number in previous generation populations is 0, crossover operation
Terminate.
Preferably in the S5 includes:
S51:Calculate each individual adaptive value;
S52:The selected probability of each individual is calculated according to adaptive value, enters next according to probability random selection is chosen
The individual in generation;
S53:If the adaptive value of the optimal solution of the new population produced is less than the adaptive value of previous generation optimal solution, on
The optimal solution of a generation replaces individual worst in new population.
Beneficial effects of the present invention are as follows:
To the improved method of genetic algorithm in being dispatched the invention provides a kind of artificial resource, it is proposed that to generating initial kind
Group optimize, the improvement of adaptive crossover operator, adaptive mutation rate, and propose be directed to restriction relation task adjust
The coded system of degree and the fitness function for reflecting individual adaptation degree by amount of unbalance, and it is proved to be effective.It is improved to lose
Propagation algorithm has stronger ability of searching optimum, and convergence is very fast, and can obtain preferable scheduling scheme, improves on the whole
The throughput of high flux emulation, not only and is randomly assigned scheme ratio excellent, relative to traditional genetic algorithm effect
More preferably.
Brief description of the drawings
The embodiment to the present invention is described in further detail below in conjunction with the accompanying drawings.
Fig. 1 shows a kind of flow of the artificial resource dispatching method specific embodiment based on improved adaptive GA-IAGA of the present invention
Figure.
Fig. 2 shows a kind of DAG figures of artificial resource dispatching method specific embodiment based on improved adaptive GA-IAGA of the present invention.
Embodiment
In order to illustrate more clearly of the present invention, the present invention is done further with reference to preferred embodiments and drawings
It is bright.Similar part is indicated with identical reference in accompanying drawing.It will be appreciated by those skilled in the art that institute is specific below
The content of description is illustrative and be not restrictive, and should not be limited the scope of the invention with this.
Task scheduling in emulation, can specifically be expressed as simulation model and system resource under multi-constraint condition on demand
Mapping problems, is the static resource allocation for realizing Simulation Application initial operating stage.The factor of influence artificial tasks scheduling has three big
Class:The communication delay between time factor and resource such as run time of restriction relation, model between model.Use heredity calculation
Method solves the problems, such as that emulation dispatch mainly needs to solve relevant parameter determination, the coding of solution, generates initial population, determines fitness letter
The problems such as number and genetic manipulation.Existing coded system is analyzed, the improved though of coding of the invention is mainly:1st, coding was both needed
Show and which resource which model is assigned in, it is also desirable to show the operation order of model in each resource.2、
The solution for ensureing generation is all effective solution, and algorithm is relatively easy.
As shown in figure 1, the invention discloses a kind of artificial resource dispatching method based on improved adaptive GA-IAGA.This method bag
Include:
S1:The population of initialization generation simulation model and artificial resource.With reference to be randomly assigned principle and optimization distribution principle
Initialize the population.The principle that is randomly assigned will be distributed to different artificial resources with mutually level simulation model
On, the optimization distribution principle matches the computing capability of the height of simulation model and artificial resource.
Specifically, when generating initial population, model computationally intensive in the group model of height identical one is distributed as far as possible
Onto the high resource of computing capability.This way takes full advantage of the characteristics of heterogeneous resource has different computing capabilitys so that
Computationally intensive model can have been run as early as possible, add the throughput of emulation.
As shown in Fig. 2 the priority constraint relationship that simulation model is reflected with DAG figures between model, each model in difference
Amount of calculation on node, and each model and which modeling communication, and the traffic between them.Represented with DAG figures
Model, each node represents a model, and the numbering of node is the numbering of model, and height of the model in DAG, reflection model exists
Priority in scheduling.Computation model is as follows in the thinking of the height of DAG figures:The height for not having the model that direct precursor puts successively is
0, otherwise, the height of the model is that the height of the node of height maximum in its direct precursor node set adds one.
Meanwhile, when generating initial population, height identical model is assigned in different resources as far as possible, realized with this
Maximum parallelization so that independent model can be run parallel, follow-up model can start to perform earlier, add emulation
Throughput, shorten simulation time.
The present invention is not complete random, except the dependence considered between model is closed when initial population is generated
System, it is also desirable to preferable initial population is generated by optimizing operation, so as to reduce the evolution time of genetic algorithm.It is initial in generation
During population, with reference to be randomly assigned with optimization distribute.Prevent fully according to pure strategy carry out the performance that distribution model is likely to result in
The excessive model of higher resource allocation so that each resource load is uneven, will also result in population and lacks diversity, therefore,
Can by certain probability stochastic assigning model.
S2:The individual fitness of each individual in population is calculated based on fitness function.Reflected according to simulation model with resource
The target penetrated determines fitness function.The target of model resource mapping of the present invention is that model is assigned to the place matched
In the resource of reason ability, while ensureing that minimum and analogue system the throughput of communication overhead is high between model.According to model resource
On the basis of the traffic between mapping objects and the amount of calculation in known models in each resource, model, the present invention is determined with negative
Amount of unbalance is carried to represent fitness function.
S3:Crossover operation is carried out to population based on single-point cross method.Crossover operation is to produce new individual in genetic algorithm
Main Means, population produced new individual by intersecting, the search space of solution extended with this, the purpose of global search is reached.
Consider that " fitness should be to relatively low crossover probability, so that outstanding individual enters of future generation higher than the individual of community average
Chance increase " thought and consider the constraint of model running order, the present invention is to single-point cross method from the following aspects
Optimize:The high individual of fitness, equiprobable selection simulation model and person's artificial resource carry out crossover operation;Fitness is low
Individual, simulation model and person's artificial resource all carry out crossover operation;The simulation model and person's artificial resource of coding use different
Cross method.Fitness is the good and bad degree for evaluating individual, and fitness is bigger, and individual is better, otherwise fitness is smaller then
Individual is poorer.The fitness of this problem refers to the load balance amount of analogue system, and load balancing techniques are to consider the distribution of isomery
The parameter such as Resource Calculation performance, communication performance in formula system is so as to ensure the effective means of all node Effec-tive Functions.Protect as far as possible
The load balance of each resource is demonstrate,proved, simulation clock could be quickly promoted, accelerates simulation process.In selection fitness height and fitness
The scope of selection is can determine that when low, for example, fitness value can be selected to be located at rear 20% positioned at preceding 20% and fitness value respectively
Individual.
The S3 may include:
S31:Two individuals are randomly choosed from population as father's individual;
S32:The adaptive value of two father's individuals is calculated, wherein larger father's individual is compared with planting group mean adaptive value;
S33:If the adaptive value of larger father's individual is less than kind of a group mean adaptive value, enter according to the probability of single-point cross method
The crossover operation of row simulation model and artificial resource, or father's individual is copied directly in population of new generation;
S34:If the adaptive value of larger father's individual is more than kind of a group mean adaptive value, enter according to the probability of single-point cross method
The crossover operation of the equiprobable selection simulation model of row crossover operation, wherein crossover operation and artificial resource, or the father is individual
Body is copied directly in population of new generation;
S35:Described two father's individuals are deleted, S31-S34 is repeated, when the individual number in previous generation populations is 0, intersected
Operation terminates.
S4:Mutation operation is carried out to population based on adaptive mutation probability.Mutation operation is that genetic algorithm produces new
Another method of body, is exactly specifically the value in some gene positions for change individual.It is the same with crossover operation, it is considered to " suitable
Response should be to relatively low crossover probability, so that outstanding individual increases into follow-on chance higher than the individual of community average
The constraint of thought and consideration model running order greatly ", the present invention uses adaptive mutation probability.
The adaptive mutation probability is the variation when population each fitness reaches unanimity or tends to local optimum
Probability becomes big;When colony's fitness is more dispersed, mutation probability diminishes.Obtained for fitness value higher than the average adaptive value of colony
Individual, takes less mutation probability;Obtain individual less than the average adaptive value of colony for fitness value, take larger mutation probability.
Fitness, which reaches unanimity, refers to that the fitness function of population is basically identical, and filial generation is reduced with parent difference, as general in do not changed variation
Rate, result of calculation will dissipate.Local optimum refers to that the fitness function of the population in certain limit is maximum, that is, load it is most balanced,
Selected according to the uniformity that the Conventional methods of selection of this area is determined.The S4 may include:
S41:An individual is randomly choosed from population as father's individual;
S42:The adaptive value of father's individual is calculated, and is compared with kind of a group mean adaptive value;
S43:If the adaptive value of father's individual is less than kind of a group mean adaptive value, emulation mould is carried out according to adaptive mutation probability
The mutation operation of type and artificial resource, or father's individual is copied directly in population of new generation;
S44:If the adaptive value of father's individual is more than kind of a group mean adaptive value, row variation behaviour is entered according to adaptive mutation probability
Make, wherein the equiprobable selection simulation model of mutation operation and artificial resource carry out mutation operation, or by father individual directly
Copy in population of new generation;
S45:Father's individual is deleted, S41-S44 is repeated, when the individual number in previous generation populations is 0, crossover operation
Terminate.
S5:Selection operation is carried out to population using based on accumulated probability improved selection opertor.The effect of selection operation is
Selecting outstanding individual to be copied directly to according to the adaptive value of individual, of future generation or producing new individual by crossover operation loses
The next generation is passed to, individual inferior is eliminated.When realizing selection operation, it is necessary to be realized according to cumulative probability.The S5 may include:
S51:Calculate each individual adaptive value;
S52:The selected probability of each individual is calculated according to adaptive value, enters next according to probability random selection is chosen
The individual in generation;
S53:If the adaptive value of the optimal solution of the new population produced is less than the adaptive value of previous generation optimal solution, on
The optimal solution of a generation replaces individual worst in new population.
The effect of selection operation is that to select outstanding individual to be copied directly to according to the adaptive value of individual of future generation or logical
Cross crossover operation and produce new individual inheritance to the next generation, eliminate individual inferior.
S6:S1-S5 is repeated, reaches and terminates after predetermined step number.
Obviously, the above embodiment of the present invention is only intended to clearly illustrate example of the present invention, and is not pair
The restriction of embodiments of the present invention, for those of ordinary skill in the field, may be used also on the basis of the above description
To make other changes in different forms, all embodiments can not be exhaustive here, it is every to belong to this hair
Row of the obvious changes or variations that bright technical scheme is extended out still in protection scope of the present invention.
Claims (9)
1. a kind of artificial resource dispatching method based on improved adaptive GA-IAGA, it is characterised in that methods described includes:
S1:The population of initialization generation simulation model and artificial resource;
S2:The individual fitness of each individual in population is calculated based on fitness function;
S3:Crossover operation is carried out to population based on single-point cross method;
S4:Mutation operation is carried out to population based on adaptive mutation probability;
S5:Selection operation is carried out to population using based on accumulated probability improved selection opertor;
S6:S1-S5 is repeated, reaches and terminates after predetermined step number.
2. according to the method described in claim 1, it is characterised in that the S1, which is combined, is randomly assigned principle and optimization distribution principle
Initialize the population;
The optimization distribution principle will be distributed to different artificial resources with mutually level simulation model, and will emulate mould
The height of type and the computing capability of artificial resource match.
3. according to the method described in claim 1, it is characterised in that the fitness function is represented with laod unbalance amount.
4. method according to claim 3, it is characterised in that the fitness function is the inverse of laod unbalance amount.
5. according to the method described in claim 1, it is characterised in that the single-point cross method is
The high individual of fitness, equiprobable selection simulation model and artificial resource carry out crossover operation;
The low individual of fitness, simulation model and artificial resource all carry out crossover operation;
The simulation model and person's artificial resource of coding use different cross methods.
6. according to the method described in claim 1, it is characterised in that the S3 includes:
S31:Two individuals are randomly choosed from population as father's individual;
S32:The adaptive value of two father's individuals is calculated, wherein larger father's individual is compared with planting group mean adaptive value;
S33:If the adaptive value of larger father's individual is less than kind of a group mean adaptive value, imitated according to the probability of single-point cross method
The crossover operation of true mode and artificial resource, or father's individual is copied directly in population of new generation;
S34:If the adaptive value of larger father's individual is more than kind of a group mean adaptive value, handed over according to the probability of single-point cross method
The crossover operation of the equiprobable selection simulation model of fork operation, wherein crossover operation and artificial resource, or father individual is straight
Connect and copy in population of new generation;
S35:Described two father's individuals are deleted, S31-S34 is repeated, when the individual number in previous generation populations is 0, crossover operation
Terminate.
7. according to the method described in claim 1, it is characterised in that the adaptive mutation probability is
When population each fitness reaches unanimity or tends to local optimum, mutation probability becomes big;
When colony's fitness is more dispersed, mutation probability diminishes;
Obtain individual higher than the average adaptive value of colony for fitness value, take less mutation probability;
Obtain individual less than the average adaptive value of colony for fitness value, take larger mutation probability.
8. according to the method described in claim 1, it is characterised in that the S4 includes:
S41:An individual is randomly choosed from population as father's individual;
S42:The adaptive value of father's individual is calculated, and is compared with kind of a group mean adaptive value;
S43:If the adaptive value of father's individual is less than kind of a group mean adaptive value, according to adaptive mutation probability carry out simulation model and
The mutation operation of artificial resource, or father's individual is copied directly in population of new generation;
S44:If the adaptive value of father's individual is more than kind of a group mean adaptive value, mutation operation is carried out according to adaptive mutation probability,
Wherein the equiprobable selection simulation model of mutation operation and artificial resource carry out mutation operation, or father individual is directly replicated
Into population of new generation;
S45:Father's individual is deleted, S41-S44 is repeated, when the individual number in previous generation populations is 0, crossover operation knot
Beam.
9. according to the method described in claim 1, it is characterised in that the S5 includes:
S51:Calculate each individual adaptive value;
S52:The selected probability of each individual is calculated according to adaptive value, enters follow-on according to probability random selection is chosen
Individual;
S53:If the adaptive value of the optimal solution of the new population produced is less than the adaptive value of previous generation optimal solution, previous generation
Optimal solution replace individual worst in new population.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710512814.7A CN107329831A (en) | 2017-06-29 | 2017-06-29 | A kind of artificial resource dispatching method based on improved adaptive GA-IAGA |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710512814.7A CN107329831A (en) | 2017-06-29 | 2017-06-29 | A kind of artificial resource dispatching method based on improved adaptive GA-IAGA |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107329831A true CN107329831A (en) | 2017-11-07 |
Family
ID=60198902
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710512814.7A Pending CN107329831A (en) | 2017-06-29 | 2017-06-29 | A kind of artificial resource dispatching method based on improved adaptive GA-IAGA |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107329831A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108320059A (en) * | 2018-02-22 | 2018-07-24 | 石家庄铁道大学 | A kind of workflow schedule evolution optimization method and terminal device |
CN109343966A (en) * | 2018-11-01 | 2019-02-15 | 西北工业大学 | A kind of cluster organization method and device of unmanned node |
CN109885401A (en) * | 2019-01-27 | 2019-06-14 | 中国人民解放军国防科技大学 | Structured grid load balancing method based on LPT local optimization |
CN111756819A (en) * | 2020-06-10 | 2020-10-09 | 赣南师范大学 | Internet of things resource scheduling optimization operation method based on improved immune heredity |
CN112463629A (en) * | 2020-12-11 | 2021-03-09 | 北京航空航天大学 | Method for adjusting software configuration items of autonomous unmanned system based on genetic evolution |
CN112612549A (en) * | 2020-12-30 | 2021-04-06 | 润联软件系统(深圳)有限公司 | Multi-edge service task selection unloading method and device and related equipment |
CN113127167A (en) * | 2021-03-18 | 2021-07-16 | 国家卫星气象中心(国家空间天气监测预警中心) | Heterogeneous resource intelligent parallel scheduling method based on improved genetic algorithm |
WO2022048280A1 (en) * | 2020-09-04 | 2022-03-10 | 苏州浪潮智能科技有限公司 | Distributed quantum computing simulation method and device |
-
2017
- 2017-06-29 CN CN201710512814.7A patent/CN107329831A/en active Pending
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108320059A (en) * | 2018-02-22 | 2018-07-24 | 石家庄铁道大学 | A kind of workflow schedule evolution optimization method and terminal device |
CN109343966A (en) * | 2018-11-01 | 2019-02-15 | 西北工业大学 | A kind of cluster organization method and device of unmanned node |
CN109343966B (en) * | 2018-11-01 | 2023-04-07 | 西北工业大学 | Cluster organization method and device for unmanned nodes |
CN109885401A (en) * | 2019-01-27 | 2019-06-14 | 中国人民解放军国防科技大学 | Structured grid load balancing method based on LPT local optimization |
CN111756819A (en) * | 2020-06-10 | 2020-10-09 | 赣南师范大学 | Internet of things resource scheduling optimization operation method based on improved immune heredity |
WO2022048280A1 (en) * | 2020-09-04 | 2022-03-10 | 苏州浪潮智能科技有限公司 | Distributed quantum computing simulation method and device |
CN112463629A (en) * | 2020-12-11 | 2021-03-09 | 北京航空航天大学 | Method for adjusting software configuration items of autonomous unmanned system based on genetic evolution |
CN112463629B (en) * | 2020-12-11 | 2022-03-29 | 北京航空航天大学 | Method for adjusting software configuration items of autonomous unmanned system based on genetic evolution |
CN112612549A (en) * | 2020-12-30 | 2021-04-06 | 润联软件系统(深圳)有限公司 | Multi-edge service task selection unloading method and device and related equipment |
CN112612549B (en) * | 2020-12-30 | 2022-06-24 | 润联软件系统(深圳)有限公司 | Multi-edge service task selection unloading method and device and related equipment |
CN113127167A (en) * | 2021-03-18 | 2021-07-16 | 国家卫星气象中心(国家空间天气监测预警中心) | Heterogeneous resource intelligent parallel scheduling method based on improved genetic algorithm |
CN113127167B (en) * | 2021-03-18 | 2023-11-03 | 国家卫星气象中心(国家空间天气监测预警中心) | Heterogeneous resource intelligent parallel scheduling method based on improved genetic algorithm |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107329831A (en) | A kind of artificial resource dispatching method based on improved adaptive GA-IAGA | |
Zhu et al. | An efficient evolutionary grey wolf optimizer for multi-objective flexible job shop scheduling problem with hierarchical job precedence constraints | |
CN110489229B (en) | Multi-target task scheduling method and system | |
CN110515735A (en) | A kind of multiple target cloud resource dispatching method based on improvement Q learning algorithm | |
CN112286677B (en) | Resource-constrained edge cloud-oriented Internet of things application optimization deployment method | |
CN110533183A (en) | The model partition and task laying method of heterogeneous network perception in a kind of assembly line distribution deep learning | |
CN106484512B (en) | The dispatching method of computing unit | |
CN104899100B (en) | A kind of resource regulating method for cloud system | |
CN104516785B (en) | A kind of cloud computing resources dispatch system and method | |
CN105975342A (en) | Improved cuckoo search algorithm based cloud computing task scheduling method and system | |
CN106227599A (en) | The method and system of scheduling of resource in a kind of cloud computing system | |
CN109542619A (en) | A kind of high-efficient carrier balance optimizing dispatching method towards cloud computing center | |
CN106250650A (en) | The resource allocation and optimization method of model in high flux emulation | |
CN107193640B (en) | Virtual machine initial placement strategy method based on Rendezvous hash algorithm | |
CN111722910A (en) | Cloud job scheduling and resource allocation method | |
CN110599068A (en) | Cloud resource scheduling method based on particle swarm optimization algorithm | |
CN108320059A (en) | A kind of workflow schedule evolution optimization method and terminal device | |
CN108681789A (en) | A kind of cloud manufacturing service optimization method | |
CN114741955A (en) | Multi-objective optimization task scheduling method based on security cloud | |
CN108399105A (en) | A kind of Method for HW/SW partitioning based on improvement brainstorming algorithm | |
CN106201706A (en) | A kind of population method for service selection and system | |
CN111309472A (en) | Online virtual resource allocation method based on virtual machine pre-deployment | |
CN116684291A (en) | Service function chain mapping resource intelligent allocation method suitable for generalized platform | |
CN113221051A (en) | Web service combination optimization method based on improved crow search algorithm | |
CN112446484A (en) | Multitask training cluster intelligent network system and cluster network optimization method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171107 |